Brian O'Neill and Alex Gillespie Simulating naturalistic instruction: the case for a voice mediated interface for assistive technology for cognition Article (Accepted version) (Refereed)

Original citation: O'Neill, Brian and Gillespie, Alex (2008) Simulating naturalistic instruction: the case for a voice mediated interface for assistive technology for cognition. Journal of assistive technologies, 2 (2). pp. 22-31. ISSN 1754-9450 © 2008 Pier Professional This version available at: http://eprints.lse.ac.uk/38691/ Available in LSE Research Online: Jan 2012 LSE has developed LSE Research Online so that users may access research output of the School. Copyright © and Moral Rights for the papers on this site are retained by the individual authors and/or other copyright owners. Users may download and/or print one copy of any article(s) in LSE Research Online to facilitate their private study or for non-commercial research. You may not engage in further distribution of the material or use it for any profit-making activities or any commercial gain. You may freely distribute the URL (http://eprints.lse.ac.uk) of the LSE Research Online website. This document is the author’s final manuscript accepted version of the journal article, incorporating any revisions agreed during the peer review process. Some differences between this version and the published version may remain. You are advised to consult the publisher’s version if you wish to cite from it.

1

Running head: Simulating naturalistic instruction Simulating naturalistic instruction: The case for a voice mediated interface for assistive technology for cognition Dr Brian O’Neill1 West of Scotland Mobility and Rehabilitation Centre Southern General Hospital 1345 Govan Road Glasgow, G51 4TF UK Tel: + 44 1412012391 Fax: + 44 1412012649 brian.oneill@sgh.scot.nhs.uk Dr Alex Gillespie Department of Psychology University of Stirling Scotland, FK9 4LA UK

1 Author for correspondence

2

Title: Simulating naturalistic instruction: The case for a voice mediated interface for assistive technology for cognition

Abstract

A variety of brain pathologies can result in difficulties performing complex

behavioural sequences. Assistive technology for cognition (ATC) attempts support of

complex sequences with the aim of reducing disability. Traditional ATCs are

cognitively demanding to use and thus have had poor uptake. A more intuitive

interface may allow ATCs to reach their potential. Insights from psychological

science may be useful to technologists in this area. We propose that an auditory-

verbal interface is more intuitive than a visual interface and reduces cognitive

demands on users. Two experiments demonstrate a novel ATC, the General User

Interface for Disorders of Execution (GUIDE). GUIDE© is novel because it simulates

normal conversational prompting to support task performance. GUIDE© provides

verbal prompts and questions and voice recognition allows the user to interact with

GUIDE. Research with non-cognitively impaired participants and a single participant

experiment with a person with vascular dementia provide support for using interactive

auditory-verbal interfaces. Suggestions for the future development of auditory-verbal

interfaces are discussed.

Keywords: Assistive technology, executive function, GUIDE©, complex behaviour,

verbal interface

3

4

Title: Simulating naturalistic instruction: The case for a voice mediated interface for assistive technology for cognition

Difficulties carrying out goal-directed behaviour lead to a high degree of disability for

which there exist few treatment options. These difficulties are manifest in groups such

as traumatic brain injury (Evans, 2003), schizophrenia (Semkovska, Bedard, Godbout,

Limoge & Stip, 2003; Krabbendam, de Vugt, Derix & Jolles, 1999) learning disability

(Cavalier & Ferretti, 1993) and the dementias, contributing to the high degree of

personal care required by persons in these groups.

In samples of those with physical and cognitive disability, use of assistive

technologies is associated with reduced need for personal assistance (Hoenig, Taylor

& Sloan, 2003). However systems designed to support cognitive function can often be

cognitively demanding, highlighting the need for systems to be useable by those with

more severe cognitive disability (LoPresti, Mihailidis & Kirsch, 2004).

Traditional scheduling technology is complex to use. Diaries require users to recall

where they are, remember to enter a prompt, remember to check the diary and monitor

task completion. Recipes schedule sub-steps of a goal state. However they also

require decoding of the written steps and information storage in working memory to

underpin performance.

The applicaion of digital technology to cognitive rehabilitation has also been limited

by the cognitive demands of interfaces. Personal digital assistants (PDAs) and

5

palmtops are designed to extend the cognitive abilities of those without impairment

and have a learning curve which places them beyond many with cognitive deficits.

Designs of simplified interfaces has allowed use by those with learning disability to

aid with time management (Davies, Stock, Wehmeyer 2002). Prospective memory

aids have benefits in reducing omissions of to-be-performed behaviours. Specifically

text prompts delivered to portable systems such as pagers and mobile phones increase

the hit rate of target behaviours (Evans, 2003). These systems are useful for those

with intact reading and direction of attention. For those with yet more severe

cognitive difficulties we suggest that systems based on auditory-verbal interfaces may

be more appropriate.

Executive Function, Language & Scaffolding

Developmental psychologists studying executive function emphasise its basis in

language, and more particularly private speech (i.e., speech directed at self rather than

an interlocutor). This developmental account was initially proposed by Vygotsky and

Luria (1930/1994; Luria,1961). Although language evolved for communication,

Vygotsky and Luria speculate that it has a secondary function enabling humans to talk

themselves through complex tasks, thus facilitating the execution of complex

purposive and planned behaviour.

Research on child development supports a relation between language and executive

function (Hughes, 1996, 2002; Barkley, 1997; Jones, Rothbart & Posner, 2003;

Zelazo, Muller, Frye & Marcovitch, 2003). Children engage in private speech when

problem solving and this increases in probability with increasing task complexity

6

(Berk & Garvin, 1984). There is also a prevalence of private speech problems

amongst children who are at risk for attention and behaviour problems (Winsler, Diaz,

Atencio, McCarthy & Chabay, 1985) and amongst children who do worse at tasks

(Frauenglass & Diaz, 1985). In adults without cognitive impairment, introducing a

secondary task which disrupts private speech disrupts performance on a planning task

(Phillips, Wynn, Gilhooly, Della Sala & Logie, 1999).

Vygotsky and Luria proposed that the child initially has minimal executive control,

and that executive control is that provided by other people, such as parents and carers.

The verbal and visual actions of these carers guide the child’s behaviour, providing a

‘scaffold’ (Berk & Winsler, 1995; Gillespie, 2006; Zittoun, Gillespie, Cornish &

Psaltis, 2007). There are activities that the child is unable to perform alone, but is able

to perform with appropriate guidance. Such activities constitute the ‘zone of proximal

development’ and scaffolding operates in this zone. The scaffold comprises mainly of

verbal guidance. Verbal guidance is used to direct attention, set sub-goals, initiate

monitoring, and correct errors. Children develop through this verbal scaffold by

internalising the guidance so that they become able to verbally guide their own

behaviour (Gillespie, 2007). The monitoring and regulation of behaviour initially

occurs between the child and the carer, but with development becomes an intra-

psychological function, namely, executive function.

The relevance of the concepts of scaffolding and the zone of proximal development

for rehabilitation have been recognised (Stone, 1998; Young et al., 2002). In the same

way that there are problems which children are unable to solve alone, but which they

are able to solve with verbal guidance, so there people with a variety of cognitive

7

deficits who confront activities of daily living which they are unable to perform alone,

but which they are able to perform with appropriate verbal guidance. We suggest that

carers are often guiding patients at the limits of the patient's ability, providing

cognitive support by using verbal prompts to scaffold patients’ executive function.

Initiation, problem-solving, generativity, planning, sequencing, organization, self-

monitoring, error correction and behavioural inhibition are functions taken on by the

carer if these are deficient in the cared-for. The basic cognitive abilities that the

patient must possess in order to gain from this instruction are language

comprehension, maintenance of a single command in mind and verbally mediated

motor programming. Carers successfully support, or scaffold, individuals with

executive dysfunction in carrying out activities of daily living by instruction (Gitlin,

et al., 2002). In this sense, carers are often acting as highly refined ‘assistants for

cognition’. Our aim has been to try and simulate the verbal scaffolding and guidance

provided by carers.

Developing an Auditory-Verbal Interface

Assistive technologies for cognition, also known as cognitive prosthetics, have the

potential to revolutionise the management of cognitive disabilities (Gregor & Newell,

2004). Prospective memory aids facilitate the performance of a behaviour at a time in

the future which might otherwise be forgotten. Traditional prospective memory aids

are commonly used and include paper notes, diaries, calendars, alarms and reminders

(Evans, Wilson, Needham & Brentnall, 2003). These examples of traditional assistive

technology have been augmented in recent years by digital assistive technology.

8

Personal digital organisers and voice recorders can now undertake the function of

several of the traditional assistive technologies, such as temporal prompting and thus

recall of the to-be-performed behaviour (Kapur, Gilisky & Wilson, 2004; Yasuda et

al., 2002). Computer systems allow central storage of schedules to be delivered as text

prompts at the point when a behaviour requires to be carried out. The effectiveness of

pager prompts as used in the proprietary Neuropage system has been demonstrated to

increase achievement of target behaviours (Wilson, Emslie, Quirk, & Evans, 2001).

Similarly the MEMEX project, utilising text messaging to mobile phones, has

demonstrated effectiveness in improving attendance at appointments and medication

compliance (Pijnenborg, Withaar, Evans, van den Bosch & Brouwer, 2007).

LoPresti, Mihailidis and Kirsch (2004) suggest that ATCs have not been achieving

their full potential in the main, due to the complexity of the ATC devices. Rather than

reducing cognitive load, they often increase the cognitive burden by requiring users

interact with complex and unfamiliar devices. Accordingly, these authors call for

future ATC devices to be more sensitive in orienting to their cognitively impaired

users (see also, Scherer, 2001).

Based on our review of the available ATCs we argue that they are biased toward

visual interfaces. Visual interfaces are attention demanding. Attentional function

predicts use of cognitive aids in sample of people with acquired brain injury (Evans,

Wilson, Needham & Brentnall, 2003). Yet the main ATCs, such as Neuropage

(Wilson, Emslie, Quirk, & Evans, 2001) and MemoJog (Inglis et al., 2002), require

users to interact with the ATC device via a screen. Users usually receive information

via the screen, and give feedback to the device via the screen. Visual interfaces are

9

good if the user’s visual system is free and the user is familiar with such interaction.

However, in cases where a user is engaged in a task (such as dressing, food

preparation, transfer or donning a limb) interacting with a computer screen entails a

shift of visual attention and an interruption of the ongoing task, thus increasing the

cognitive load of the task. Where users are unfamiliar with computers, the device will

also be unintuitive contributing to low uptake.

We argue, on the basis of the literature on scaffolding, that an effective way forward

is for ATCs to model the cognitive support provided by carers, or ‘assistants for

cognition’. First, the cognitive scaffolding provided by carers is verbal, not visual, and

thus does not lead the patients’ visual attention away from the task at hand. Second,

given the evidence presented about the close relationship between executive function

and language, we speculate that prompting in the verbal medium rather than the visual

medium provides a more direct augmentation of executive function. Third, the verbal

guidance provided by carers is in a familiar mode of interaction, namely,

communicative interaction, and thus there is no learning curve. Finally, the

scaffolding provided by carers is task-focused and tailored to the individual, that is to

say each verbal prompt is contextually relevant to the immediate sub-goal that the

user is engaged in. Written instructions, flow charts or diagrams, on the other hand,

often present all the prompts and guidance at the same time, for example on the one

sheet of paper or flow chart. Such information overload is avoided in the auditory

medium because the linearity inherent in the auditory medium ensure that only one

prompt is presented at a time.

10

GUIDE©

We have developed an ATC that simulates the type of guidance provided by carers.

The device is called GUIDE© - General User Interface for Disorders of Execution.

The GUIDE© uses the most generalizable interface known for guiding users through

complex tasks, namely verbal guidance. The GUIDE© prompts users, asks users

questions and accepts verbal responses. The GUIDE© uses the verbal responses to

direct the deployment of subsequent prompts and questions. Delivering prompts and

questions to users verbally (rather than visually) entails less cognitive load, and does

not require users to switch task-focus (such as shifting from task to viewing a screen

or a card). The GUIDE© is designed to augment task focus, not to interrupt it. The

GUIDE© does not disempower users, by undermining their agency, rather the

GUIDE© empowers users. The GUIDE© prompts users with simple questions,

requiring the user to engage in the task in order to answer. The range of answers that

the GUIDE© accepts is deliberately limited, so as to reduce cognitive load. An

assumption is that verbal responses are comparatively easy for users, requiring the

least cognitive load and are similar to verbal interactions with carers

The GUIDE© is in the tradition of COACH (Mihailidis, Barbenel & Fernie, 2004) in

that it is focused upon a particular task and uses verbal prompts. The GUIDE©

incorporates Mihailidis, Barbenel and Fernie’s (2004, p. 166) recommendations for

constructing a verbal prompting system: prompts are provided at different levels of

detail, prompts are tailored to the individual user, prompts only pertain to one action

at a time, and prompts are phrased in a terminology that is familiar. The GUIDE©

also goes beyond these recommendations by virtue of attempting explicitly to

11

simulate the scaffolding support provided by carers. The prompts used by the

GUIDE© are modelled on the prompts provided by carers. Users interact with the

GUIDE© in the same way that they do with carers, namely, verbal exchange.

Specifically, this means that many of the prompts are actually questions, and these

questions require a verbal response from the user. The questions engage the user and

are intended to stimulate and augment, rather than command.

Research on a range of ATCs has tended to conclude that generic systems come into

problems both due to the particularities of the task to be achieved and particularities

of the patient (Cole, 1999; Mihailidis, Barbenel & Fernie, 2004). In order to avoid

such problems protocols are carefully constructed to address the task to be achieved

by the user.

Technologically, the GUIDE© has four components: hardware, voice recognition

software, an action pathway used to ‘scaffold’ users behaviour, and the GUIDE©

software program, running on a PC, that coordinates the parts. The action pathway

module within the GUIDE© stores steps and checks based on observations of

competent performance of the task. Verbal prompts are recorded and stored on the

PC. Verbal responses are processed via voice recognition software.

In order to develop our argument that audio-verbal interfaces are particularly suited to

ATCs which target behavioural sequences which require visual attention, we present

data from two recent studies using GUIDE© (Dickie, Tyler, O’Neill & Gillespie,

submitted; O’Neill & Gillespie, submitted). The first presents the results of a study of

adults without cognitive impairment carrying out a novel task with a distracting

12

secondary task comparing GUIDE use with following written instructions. The

second presents a case study of a man with vascular dementia carrying out a self-care

behaviour which he failed to acquire in rehabilitation as usual.

Study 1: Using GUIDE© with non-impaired adults

Design: A between-participants design was used with participants randomly allocated

to two groups. The task was to make an elaborate smoothie. One group received only

written instructions while the other group received the same instructions via the

GUIDE©. Periodic distraction was simulated in both conditions with a random

number generation task.

Participants: 26 participant sample of convenience (mean age 26; range19 to 43; 17

males, 9 females) randomly allocated to one of two groups. The Written Instruction

Group comprised 10 male and 3 female. The GUIDE© Group comprised 7 male and

6 female. No significant age differences existed between the groups.

Procedure: Before beginning the task participants were trained in the use of GUIDE©

for about 10 minutes. Training of the voice recognition software also occurred at this

time. The five commands were repeated 15 times each to reach an adequate level of

recognition. The task occurred in a kitchen with all equipment and ingredients

present. During the smoothie making task, participants were asked to generate a

random number between 34 and 43 every 30 seconds, throughout the experiment.

13

Scoring: Three indices of performance were independently rated: Percentage of steps

completed successfully; frequency of hesitations, other than when reading or listening

to prompts, of greater duration than 10 seconds; and mean time per step of the task.

Inter-rater reliability, measured with Pearson’s correlation coefficient was high for

number of correct steps (r=0.92, p<.001), for deviations (r=0.95, p<.001) and for

hesitations (r=.73, p<.001).

Following completion of the task, participants completed a questionnaire designed to

assess the participants’ feelings towards the method of instruction they experienced,

whether written instructions or the Guide© device. This questionnaire was scored by

counting the total number of positive and negative comments reported by the

participant.

Results: Table 1 shows the group means for percentage of correct steps and incorrect

steps, number of hesitations and average time per step.

[insert Table 1 about here]

For the number of steps completed correctly, the Guide© performed significantly

better than the written instructions (F(1) = 6.375, p<0.05). The number of hesitations

show by users of the Guide© was significantly fewer than those shown by the group

following written instructions (F(1) = 12.084, p<0.05). The average time per step was

not significantly different, however there was a trend towards the Guide© steps being

completed more quickly than those in the written instruction group (F(1) = 1.326,

p>0.05). The mean number of positive comments given in the questionnaires was 2.5

14

for the Guide© (SD 1.31) and 1.0 for the written instructions (SD 1.04). This was

significantly different (F(1) = 9.581, p<0.05).

Study 2: Using GUIDE© with a cognitively impaired user

Participant: A 67 year old man (BB) who underwent bilateral transtibial amputations

due to peripheral arterial disease and comorbid diabetes a year prior to this

intervention. Following amputation he had received out-patient rehabilitation in a

specialist centre for post-amputation rehabilitation three times weekly for 8 weeks.

This consisted of instruction from a physiotherapist on the correct donning of his

prosthetic limbs and mobilization using a variety of supportive walking aids. He

consistently made errors in donning his prosthetic limbs and omitted critical safety

checks. He was thus advised by his physiotherapists that he was not to don his

prostheses without supervision. A standard neuropsychological assessment

(RANDOLPH, 1998) demonstrated impairments across several cognitive domains

with visuospatial function, attention and delayed memory lying in the extremely low

range.

Provision of visual sequence prompts (a series of client-viewpoint photographs of

correct steps) was tried for six limb donning attempts. He was unable to benefit from

these. We suspect that this was due to the attentional demands of shifting attention

from current task to the visual cues.

Task: A standardized sequence was developed for prosthetic limb donning, following

interview with physiotherapists and prosthetists. This was then instantiated as a

15

computer-stored sequence GUIDE© and delivered to the patient via wireless

headphones.

Design: A single participant baseline-intervention experiment was employed.

Baseline data were gathered by video recording on-task behaviour. These videos were

scored, using a customised Sequence Performance Scale, for a. percent correct b.

errors of omission, c. deviations from sequence, d. repetitions and e. time to fit limb to

give data points for statistical analysis (Semkovska et al., 2003). Inter-rater reliability

for this measure ranged from 0.73 to 0.77 across the indices of the measure. The

intervention trials were randomly allocated within the sequence of trials.

Results: The participant’s performance scores by trial are shown in Figures 1 and 2.

On the x axis one can see the trial number, separated into baseline on the left and

intervention on the right. BB was on average 81% correct in the baseline and 94%

correct on the intervention trials. An exact probability test for randomized trials

(Todman & Dugard, 2001) found that GUIDE© use was associated with statistically

significant reductions in repetitions of previous steps (p<0.05) and omissions of

sequence steps (p<0.05). Reduction in deviations from the ideal sequence approached

significance (p=0.061). Time to don the prostheses was reduced from a mean of 12

minutes in baseline to 9.25 minutes on GUIDE© trials. An Efficiency Index (Total

Correct / Total Time) also changed significantly between conditions (p<0.05).

[Insert Figures 1 about here]

[Insert Figure 2 about here]

16

Discussion

Both of our preliminary studies show benefits of using a voice mediated prompting

system over sequence performance without. This was in the context of imposed

distractors in Study 1 and in a participant with cognitive impairment in Study 2.

Taken together, these studies show the feasibility of using an auditory-verbal interface

in ATC design for contexts of high distraction and cognitive impairment.

The GUIDE© demonstrates potential as a rehabilitation tool. Cognitive problems

compromise rehabilitation outcome in conditions such as cerebrovascular accident

(Paolucci et al., 1996) and post-amputation (O’Neill, 2008). For many of these

conditions rehabilitation is labour intensive due to the need for repeated instruction or

supervision to prevent errors. Computer assisted guidance of repeated instruction

could thus allow patients to engage autonomously in rehabilitation relevant tasks.

The success of the GUIDE©, we suggest, stems from the verbal interface. While the

mode of verbal interaction is novel for ATCs, it is familiar for users. BB adapted to

the use of the GUIDE© in the first session. While many other ATCs have reported

difficulties in training users and long learning curves (LoPresti, Mihailidis & Kirsch,

2004), we suspect this has been due to the lack of familiarity with the interface. A

second advantage with the verbal interface, which we argued in the introduction, is

that it provides a relatively direct route to augmenting executive function because

there is evidence to suggest that executive function is in fact heavily verbally

17

mediated. The success of the GUIDE© is consistent with such an association.

Moreover, during the trials when BB was not using the GUIDE© the researchers

noted an increased tendency for BB to talk himself through the task. This talk was

loud enough to be heard by the researchers. Such utterances are in line with research

which has shown that children facing a complex task tend to talk themselves through

the task (Berk & Garvin, 1984; Vygotsky & Luria, 1994; Winsler & Naglieri, 2003),

thus again supporting the link between executive function and verbal mediation which

the GUIDE© is based upon.

All users in the above studies showed good tolerance of the system. In Study 1 more

positive comments were made with regard the GUIDE© than the written instructions.

It is hypothesised that a verbal guidance system is tolerated because the delivery of

prompts is similar to the experience of private speech, which underpins normal

problem solving (Phillips, Wynn, Gilhooly, Della Sala & Logie, 1999). BB

commented after some trials using the GUIDE© ‘how did I do?’ or ‘I did well there’.

This contrasted with his previous thanking of the instructor or making disparaging

comments about them in the event of poor performance. It thus appears likely that

persons using the system may be able to attribute success to the actions of the self

rather than use of the system. This is promisingly indicates that users may experience

personal satisfaction attaining a goal using the system, despite being scaffolded to do

so.

The acceptability of having a sequence step cued by a verbal stimulus appears good.

What is less clear is the acceptability of making verbal responses to control cue

deployment. Each of the participants performed the focal task alone apart from the

18

presence of the experimenter and it remains a possibility that users may feel self-

conscious if using the system in earshot of others. Future research is needed to

examine the extent to which using the device in public settings is perceived as

stigmatising.

The emulation of carers’ verbal guidance of behaviour in the field of assistive

technology for cognition is rare. Orpwood, Adlam, Gibbs and Hagan (2001) used

recorded carers’ voices to remind persons with dementia of omissions triggered by

sensors. For example a bath sensor triggered the recording: “Don’t forget you’ve left

the bath running mum”. Literature search did not elicit any research assessment of the

effectiveness of such devices although the initial case studies are promising. To our

knowledge our system is the first to accept users’ verbal responses to control output.

We would argue that some control of the output is an important feature. Being unable

to control the voice directing behaviour may be noxious and disempowering.

Perception of an assistive technology is an important determinent of use (e.g. Scherer,

2002) and we argue that the acceptability of devices (e.g. monitoring systems,

prompters etc.) should be to the fore of any assessment of their efficacy.

The studies reported here demonstrate the orthotic function of the system.

Performance-with was only compared to performance-without and there was no

examination of learning effects. However, for individuals with better memory

function the system might also restore performance of complex sequences. Errorless

learning, wherein performance errors are minimised for the memory impaired learner,

has evidence to suggest efficacy in teaching ordered semantic information (Kessels &

De Haan, 2003) and behavioural sequences (Maxwell, Masters, Kerr & Weedon,

19

2001). The system can guide error free performance and might thus be used to support

errorless learning. Directions can be gradually replaced by question prompts with

these faded out in turn, incrementally increasing the amount to be recalled.

The range of potential applications is difficult to estimate but appears wide. All

complex behavioural sequences which can be verbally described are open to the

application. For future research we suggest identifying and then augmenting those

behaviours which formal and informal carers are currently providing verbal

scaffolding for. In depth analysis of carer providers as ‘assistants for cognition’ might

provide further clues as to the most appropriate modes of questioning and prompting.

Also studies of the self-talk that normally accompanies the prefromance of a complex

behavioural sequence are needed. The motivation for this research is to construct the

simplest and most intuitive interface for ATCs. To this end, the future development of

ATC interfaces, we suggest, should focus upon the auditory-verbal interface to

simulate both carer instruction and the internal verbal control of sequence

performance.

References

Barkley R A (1997) ADHD and the nature of self-control. New York: The Guilford

Press.

Berk L E & Garvin R A (1984) Development of private speech among low-income

Appalachian children. Developmental Psychology 20 271-286.

20

Berk L E & Winsler A (1995) Scaffolding children's learning: Vygotsky and early

childhood education. Washington: National Association for the Young Children.

Cavalier A R & Ferretti R P (1993) The use of an intelligent cognitive aid to facilitate

the self management of vocational skills by high school students with severe learning

disabilities. In RESNA'93 Annual Conference. Arlington, VA: RESNA Press.

Cole E (1999) Cognitive prosthetics: An overview to a method of treatment.

NeuroRehabilitation 12 39-51.

Crepeau F & Scherzer P (1993) Predictors and indicators of work status after

traumatic brain injury: A meta-analysis. Neuropsychological Rehabilitation 3 5-35.

Davies D, Stock S, Wehmeyer M (2002) Enhancing independent time-management

skills in those with mental retardation using a palmtop personal computer. Mental

Retardation 40 (5) 358-65

Evans J J (2003) Rehabilitation of executive deficits. In B. A. Wilson (Ed.)

Neuropsychological Rehabilitation. Abingdon: Swets and Zeitlinger.

Evans J J, Wilson B A, Needham P, & Brentnall S (2003) Who makes good use of

memory aids? Results of a survey of people with acquired brain injury. Journal of the

International Neuropsychology Society 9 925-935.

21

Frauenglass M H & Diaz R M (1985) Self-regulatory functions of children's private

speech: A critical analysis of recent challenges to Vygotsky's theory. Developmental

Psychology 21 357-364.

Gillespie A (2006) Games and the development of perspective taking. Human

Development 49 87-92.

Gillespie A (2007) The social basis of self-reflection. In J. Valsiner & A. Rosa (Eds.),

The Cambridge Handbook of Sociocultural Psychology. Cambridge: Cambridge

University Press.

Gitlin L, Winter L, Dennis M, Corcoran M, Schinfeld S & Hauck W W (2002)

Strategies used by families to simplify tasks for individuals with Alzheimer's disease

and related disorders: Psychometric analysis of the Task Management Strategy Index

(TMSI). The Gerontologist 42 61-70.

Gregor P & Newell A (2004) Introduction. Neuropsychological Rehabilitation 14 1-3.

Hoenig H, Taylor D & Sloan F (2003) Does assistive technology substitute for

personal assistance in the disabled elderly? American Journal of Public Health 93 (2)

330-337.

Hughes C (1996) Control of action and thought: Normal development and

dysfunction in autism: A research note. Journal of Child Psychology and Psychiatry

37 229-236.

22

Hughes C (2002) Executive functions and development: Emerging themes. Infant and

Child Development 11 201-209.

Inglis E, Szymkowiak A, Gregor P, Newell A, Hine N, Shah P, Evans J J & Wilson B

A (2003) Issues surrounding the user-centred development of a new interactive

memory aid. Universal Access in the Information Society 2 226-234.

Jones L B, Rothbart M K, & Posner M I (2003) Development of executive attention in

preschool children. Developmental Science 6 (5) 498-504.

Kapur N, Glisky E L, & Wilson B A (2004) Technological memory aids for people

with memory deficits. Neuropsychological Rehabilitation 14 41-60.

Kessels R P C & de Haan E H F (2003) Implicit Learning in Memory Rehabilitation:

A Meta-Analysis on Errorless Learning and Vanishing Cues Methods. Journal of

Clinical and Experimental Neuropsychology 25 805-814.

Krabbendam L, de Vugt M E, Derix M M & Jolles J (1999) The behavioural

assessment of the dysexecutive syndrome as a tool to assess executive functions in

schizophrenia. Clinical Neuropsychology 13 370-375.

Lezak M D (2003) Neuropsychological Assessment, 4th Edition. Oxford: Oxford

University Press.

23

LoPresti E, Mihailidis A, Kirsch N (2004) Assistive Technology for Cognitive

Rehabilitation: State of the Art. Neuropsychological Rehabilitation 14 (1/2) 5-39.

Luria A (1961) The role of speech in the regulation of normal and abnormal

behaviour. London: Pergamon Press.

Maxwell J, Masters R, Kerr E & Weedon E (2001) The implicit benefits of learning

without errors. The Quarterly Journal of Experimental Psychology 54 1049-1068.

Mihailidis A, Barbenel J C & Fernie G (2004) The efficacy of an intelligent cognitive

orthosis to facilitate handwashing by persons with moderate to severe dementia.

Neuropsychological Rehabilitation 14 135-171.

O’Neill, B (2008) Cognition and Mobility Rehabilitation following Lower Limb

Amputation. In: Gallagher, Desmond and MacLachlan (Eds.) Psychoprosthetics:

State of the Knowledge. London: Springer.

Orpwood R, Adlam T, Gibbs C & Hagan S (2001) User-centred design of support

devices for people with dementia for use in a Smart House. In Marincek C, Buhler C,

Knops H & Andrich R (Eds) Assistive Technology –Added Value To The Quality of

Life. Amsterdam: IOS Press.

Paolucci S, Antonucci G, Guariglia C, Magnotti L, Pizzamiglio L & Zoccolotti P

(1996) Facilitatory effect of neglect rehabilitation on the recovery of left hemiplegic

stroke participants: A cross-over study. Journal of Neurology 243 308-396.

24

Phillips L, Wynn V, Gilhooly K, Della Sala S, & Logie R (1999) The role of memory

in the tower of London task. Memory 7 209-231.

Pijnenborg G H, Withaar F K, Evans J J, van den Bosch R J, & Brouwer W H (2007)

SMS text messages as a prosthetic aid in the cognitive rehabilitation of schizophrenia.

Rehabilitation Psychology 52 236-240

Scherer M J (2001) Assistive technology: Matching device and consumer for

successful rehabilitation. Washington: American Psychological Association.

Semkovska M, Bedard M A, Godbout L, Limoge F & Stip E (2003) Assessment of

executive dysfunction during activities of daily living in schizophrenia. Schizphrenia

Research 69 289-300.

Shallice T (1982) Specific Impairments of Planning. Philosophical Transactions of

the Royal Society of London B 298 199-209.

Singer B D & Bashir A S (1999) What are executive functions and self-regulation and

what do they have to do with language-learning disorders? Language, Speech, and

Hearing Services in Schools 30 265-273.

Stone C A (1998) The metaphor of scaffolding: Its utility for the field of learning

disabilities. Journal of Learning Disabilities 331 344-364.

25

Tierney M, Snow W, Charles J, Moineddin R & Kiss A (2007) Neuropsychological

predictors of self-neglect in cognitively impaired older people who live alone. The

American Journal of Geriatric Psychiatry 15 140-148.

Todman J B & Dugard P (2001) Single Case and Small-n Experimental Designs: A

Practical Guide to Randomisation Tests. London, Lawrence Erlbaum Associates.

Vygotsky L S & Luria A (1930/1994) Tool and symbol in child development. In Van

de Veer R & Valsiner J (Eds.) The Vygotsky Reader. Oxford: Blackwell.

Wilson B A, Emslie H C, Quirk K & Evans J J (2001) Reducing everyday memory

and planning by means of a paging system: a randomized control crossover study.

Journal of Neurology Neurosurgery and Psychiatry 70 477-482.

Winsler A, Diaz R M, Atencio D J, McCarthy E M & Chabay L A (2000) Verbal self-

regulation over time in preschool children at risk for attention and behavior problems.

Journal of Child Psychology and Psychiatry 41 875-886.

Winsler A & Naglieri J (2003) Overt and covert verbal problem-solving strategies:

Developmental trends in use, awareness and relations with task performance in

children aged 5 to 17. Child Development 74 659-678.

Woodruff-Pak D S (1997) The Neuropsychology of Ageing. Oxford: Blackwell.

26

Yasuda K, Misu T, Beckman B, Watanabe O, Ozawa Y & Nakamura T (2002) Use of

an IC recorder as a voice output memory aid for patients with prospective memory

impairment. Neuropsychological Rehabilitation 12 155-166.

Young D A, Zakvanis K K, Campbell Z, Freyslinger M G & Meichenbaum D H

(2002) Scaffolded instruction remediates Wisconsin Card Sorting Test deficits in

schizophrenia: A comparison to other techniques. Neuropsychological Rehabilitation

12 257-287.

Zelazo PD, Muller U, Frye D & Marcovitch S (2003) The development of executive

function in early childhood. Monographs of the Society for Research in Child

Development 68 vii-137.

Zittoun T, Gillespie A, Cornish F & Psaltis C (2007) The metaphor of the triangle in

theories of human development. Human Development 50 208-229.

Tables and Figures

Table 1 – Mean and Standard Deviation Scores for Key Measures

The Guide ©(n=13) Written Instructions (n=13)

Measure M SD M SD

Percentage of Correct Steps 91.03 12.16 74.83 19.67

Percentage of Incorrect

Steps

8.97 12.16 25.17 19.67

Number of Hesitations 1.35 0.99 2.96 1.55

27

Mean time for each Step 1.15 0.29 1.29 0.29

Positive Comments 2.50 1.31 1.00 1.04

Negative Comments 1.50 1.31 1.50 1.51

Figure 1: Sequence performance repetitions by condition

0

1

2

3

4

5

6

7

8

1 3 4 6 9 2 5 7 8

Trial

Freq

uenc

y

Repetitions

Baseline Intervention

Figure 2: Sequence performance omissions by condition

28

0

1

2

3

4

5

6

7

8

1 3 4 6 9 2 5 7 8

Trial

Freq

uenc

y

Omissions

Baseline Intervention